This project has established a series of core neuro-biological findings regarding the basis of Conduct Disorder (CD), particularly CD with Callous-Unemotional traits (reduced guilt and empathy). Over the past year, our main achievements have been the following: First, in earlier work on this protocol, we demonstrated that CD+CU is associated with impaired processing of the fear of others and this impairment is associated with a reduced response by the amygdala to this expression. This year we contrasted two prominent explanations for this deficit: (a) a fundamental reduction in responsiveness of the amygdala to this expression (the emotion position); (b) an over-responsive top down attentional system directing attention to non-emotional features of the stimulus and thus reducing the emotional response (the attention position). Our work revealed support for the emotion position. Youth with CD+CU showed appropriate (i.e., not enhanced) recruitment of systems implicated in top down attentional control and group differences in amygdala responses to fear were only observed under low attentional load conditions. Moreover, in a second study, we demonstrated that attention effects seen in CD+CU appeared to driven by their emotional deficits. Thus reductions in attention orienting (and the recruitment of neural systems mediating this orientation) were selective for fearful expressions. Second, in earlier work on this protocol, we demonstrated that CD+CU is associated with impaired reward/punishment based decision making and this impairment related to dysfunction within orbital frontal cortex and caudate. However, our previous work did not involve computational modeling of the neural response data and thus although we knew these regions were dysfunctional we did not know how they were dysfunctional. In particular, there are two core processes critical for successful decision making. The first of these is prediction error signaling; i.e., signaling than an outcome is better or worse than the individual expected it to be. This is critical because the more surprising the outcome, the more the individual needs to alter their reward/punishment expectation of the action that gave this outcome. The second is expected value; i.e., representing the likely reward or punishment that will be received following performance of the action. This is critical because the poorer the individual is at predicting future rewards/ punishments, the poorer the decisions they will make. In work conducted over the past year, we demonstrated that both of these processes are impaired in youth with CD+CU. These youth show profound impairment in prediction error signaling within caudate and significant impairment in the representation of expected value within orbital frontal cortex and the insula. In further work, we showed atypical responding in these regions in the context of a social exchange task relate specifically to the individuals level of CU traits. Critically, this work has allowed us to identify objective bio-markers that can be used in on-going work to assess treatment efficacy in this population.